Message
Re: [colorvision_group] Questions about measuring targets
2007-04-24 by CDTobie@aol.com
In a message dated 4/24/07 9:45:21 AM, gene51@... writes:
I'm using a Canon i9100 and Printfix Pro Suite, current version. I am
having some difficulty profiling, but I also have some basic
questions as to what to expect.
I've printed a 729 patch three page target, making sure there are no
settings producing any color management in the printer driver, and
did the test for the best media settings.
Thats certainly the slow way to do your testing. Its most reasonable to work with tthe 225 patch targets, at least until you have everything figured out.
The first question. Shouldn't the measured target screen view
somewhat resemble the colors in the printed target?
Not really, one is raw RGB color on screen, the other is measured print colors from an unknown printer, ink and paper combination... generally speaking printer output will be signficantly darker.
The colors are
all off, and in most cases darker than the printed target.
Sounds right...
For
example, on the third page, row 14, columns G through I show gray
patches on the screen view, yet the printed target has very light
yellow.
I wouldn't worry about this until after I'd built a profile and tested it...
The final profile has a distinct red cast, and is darker and more
contrasty than the on screen view. The gray ramp looks ok as far as
gray values are concerned, but their is a section that has a very
light red cast in the lighter area, and further down there is a
slight yellow cast. How do you get rid of this?
So you are seeing two color casts in grays, and an overall cast? That could mean you are simply doing something wrong in the process, or it could be related to metamerism, but its sounds more like process error...
In the test image, the shadows have a reddish cast, and the area
behind the sunflower also has a red cast. Adding cyan doesn't remove
it completely. But the question here is shouldn't the image after
profiling be reasonably close to the screen image? Isn't that the
whole purpose of profiling the monitor and the printer?
I'd suspect patch reading issues. Start by printing and carefully reading a 225 patch target, using minimal margins to get good big patches. If that target does not produce nearly identical print results, that would indicate that one or the other or both profiles were done incorrectly.
I hope I am doing something wrong, so I can fix it by improving my
technique.
That would be my first guess...
I have gotten good results with kodak premium plus glossy,
admittedly not a great paper, and epson premium glossy photo - but I
feel that it was more by accident than by design. In each case, the
first printed results after profiling had contrast and color casts,
which I was able to remove with the tweaks in Printfix Pro.
Shouldn't have to do that... certainly not to any significant degree. You would need to send me (me, not the list) detailed info about your process, print sizes, copies of your measurement XML files, etc, for me to troubleshoot this.
C. David Tobie
Product Technology Manager
ColorVision Business Unit
Datacolor Inc.
CDTobie@...
www.colorvision.com
**************************************
See what's free at http://www.aol.com.